The endothelium is highly sensitive to acetylcholine (ACh) and responds to the neurotransmitter by releasing vasodilators that govern underlying smooth muscle tone. However, ACh released by neurons is not expected to reach the endothelium through the vessel wall or via blood. Despite a lack of a clear physiological role, the endothelium is so exquisitely sensitive to ACh that the neurotransmitter is the most frequently used assay of endothelial function. Here, by studying calcium (Ca2+) signalling in large areas of the endothelium of intact arteries, we show that the endothelium itself is a source of ACh and that ACh underlies flow-mediated vascular responses. Ca2+ signalling was studied in large fields of endothelia (~150 cells) of common carotid and second-order mesenteric arteries from male (Sprague-Dawley, 150-250g) rats killed by overdose of CO2 (Schedule 1; Animals (Scientific Procedures) Act 1986). In both carotid and mesenteric arteries, flow evoked repeatable, complex endothelial Ca2+ signalling. Flow-evoked responses persisted in Ca2+-free physiological saline solution and were sensitive to 2-aminoethoxydiphenyl borate (100µM), cyclopiazonic acid (100µM), and U71322 (5µM), suggesting the Ca2+ rises derived from phospholipase C-dependent, InsP3-induced Ca2+ release form internal stores. Importantly, flow-mediated Ca2+ responses could be manipulated by modulating cholinergic signalling pathways: the muscarinic receptor blocker, atropine (100nM), and the ACh hydrolase, acetylcholinesterase (AChE; 4U/ml), each abolished flow-induced Ca2+ activity, whilst the AChE inhibitor, neostigmine (10µM), enhanced Ca2+ activity, and the choline acetyltransferase inhibitor, bromoacetylcholine (50µM), attenuated flow-evoked responses. Ca2+ responses were unaffected by the vesicular ACh transporter inhibitor, vesamicol (10μM) but were inhibited by corticosterone (100μM), and abolished by decynium-22 (1μM), inhibitors of organic cation transporters (OCTs). These results suggest that non-vesicular release of ACh occurs via OCTs. Ca2+ responses were unaffected by the ATP hydrolase, apyrase (4U/ml), the P2 receptor antagonist, suramin (100µM), or the pannexin-1 blocker, probenecid (250µM), suggesting that ATP release does not contribute to the flow response. Flow-evoked signals were also insensitive to tetrodotoxin (10 μM) suggesting that nerves are an unlikely contributor to the response. All treatments tested in the same preparation at multiple time points, and responses compared using one-way ANOVA with Dunnet’s post hoc test (n≥3; p<0.01 considered significant). These results uncover a new role for the classical neurotransmitter, ACh, in endothelial mechanotransduction, identify the endothelium as a source of non-neuronal ACh release by fluid flow and thus demonstrate a physiological reason for the endothelium’s sensitivity to ACh.